108                           ALDO LEOPOLD AND Z. 
 
  A conspicuous character of this group is the early 
bloom and short duration of bloom in the first 21 
species. Their average duration is 21 days. These 
species bloom before the leafing of trees is com- 
plete, and thus contrive to receive at least partial 
sunlight, even when occupying sites shaded by over- 
head trees, and often by a shrubby understory as 
well. That this short early bloom is actually an 
evolutionary adaptation for seizing sunlight is sus- 
tained in some species by the fact that the leaves 
turn yellow and die as soon as fruiting is completed. 
For example, the leaves of Dutchman's breeches and 
dogtooth violet are yellow by the end of May and 
dead by mid-June. This early death occurs even in 
transplanted individuals with plenty of light and 
moisture. The yellowing foliage of these two species 
is the earliest "fall color" of the year. 
             PRAIRIE AND SAND PLANTS 
   The average blooming dates of 59 forbs and grasses 
are assembled in Figure 2. Of the 14 flowers which 
bloom before June 1, nine are confined to hot dry 
sands and gravels. On such sites, in very drouthy 
springs, snow water is the only moisture likely to 
be available. Pasque [44] and Draba [51] are ex- 
amples of sand plants which can bloom on snow 
water. 
   In some early sand flowers, the blooming period 
 is short regardless of the weather; thus Draba did not 
 span more than two weeks in any year. Other early 
 species are opportunist; thus Arabis lyrata [74] 
 bloomed from April 11 to June 6 and straggled to 
 August 12 during the cool spring of 1945, whereas 
 in normal years it ceases to bloom in May. 
   Some prairie grasses and forbs start the season's 
 growth very late. Could this be an evolutionary de- 
 vice for avoiding damage from spring fires? For 
 example, bluestem   [290] grows no visible leaves 
 until late May, whereas most other perennial grasses 
 become green in April. Butterfly weed [241] does 
 not sprout until June 1, whereas the other milk- 
 weeds sprout a month earlier. 
   The prairie group is peculiar in its interspersion 
 of long and short blooming plants. Its long collec- 
 tive span, together with the tendency of prairie 
 grasses to color in fall, gives it great value in wild 
 landscaping. Every month from April to October 
 offers something to see on a prairie. 
                       WEEDS 
   The average blooming dates of 46 common weeds 
 are presented in Figure 3.     As has often been 
 pointed out, there is no objective definition of the 
 term "weed," and none is claimed for Figure 3, ex- 
 cept that it includes plants which somebody, for some 
 .reason, dislikes. The group might well be larger: 
 thus Asiatic honeysuckle [123] is probably a greater 
 threat to the native flora in this region than 
 any weed in Figure 3, barring quack grass [197]. 
 Of the 46 species, not more than half a dozen are 
 dangerous in the sense of usurping land that has not 
 been abused. Most Wisconsin weeds are objection- 
 ably abundant mainly in overgrazed pastures [239, 
 
 
A 
 
 
RA ELIZABETH JONES                Ecological Monographs 
                                         Vol. 17, No. 1 
255, 259, 282, 289, 298, 307], and are preferable 
to the erosion which would be augmented by their 
absence. Winter animals are largely dependent on 
weed seeds. 
   The exotic perennials which comprise the bulk of 
 the weed list have certain phenological characters in 
 common. First, their period of bloom is long. The 
 average duration, exclusive of straggling, is 45 days. 
 By reason of straggling, the termination of bloom 
 often lacks sharpness. Second, they commonly re- 
 sprout and rebloom after mowing, and also after 
 interruption of bloom by drouths, Examples: [53, 
 182, 255]. Some display straggling blooms even 
 when not interrupted by mowing or drouths. Many 
 straggle until frozen. Examples: [177, 218, 255]. 
                  HAYFEVER WEEDS 
   There is a discrepancy between the development 
 of pollen in the two ragweeds at the Dane station, 
 and the detection of aerial pollen by the Allergy 
 Laboratory of the Wisconsin General Hospital. In 
 two of the last three years the hospital has detected 
 pollen in the air before any phenologist could detect 
 general bloom in the field. 
   There is also a discrepancy between first general 
 bloom in ragweed, and first pollen in a few preco- 
 cious plants. In two of the last three years one or 
 more precocious individual plants have been found, 
 by accident, far ahead of the general population: 
 First         Precocious         General  -Allergy 
 Pollen         Plants           Population Laboratory 
 1944 -----......... x              7/30      7/28 
 1945 ................ 7/13, 8/1  (both  lesser)  8/7  8/1 
 1946 ..---------- . 7 /26  (greater)  7/31   7/31 
 Average difference..................15 days..-........... 3 days 
   The question is: does the earliest pollen detected 
 by the Allergy Laboratory originate in other regions, 
 or does it originate in these precocious local plants? 
 The present data yield no clue. 
   The precocious plants are very scarce, and are 
 doubtless genetical aberrations. Three years' search 
 has yielded only the three plants. 
   The average Dane County dates (August 5 for 
 lesser and August 8 for greater ragweed) are later 
 than the August 2 average for both species in this 
 region given by Wodehouse in "Hayfever Plants" 
 (1945). 
                    MARSH PLANTS 
    The average blooming dates of 28 marsh plants 
 are presented in Figure 4. This group consists, in 
 large part, of conspicuous abundant perennials with 
 long blooming periods (average duration 38 days). 
 As in the prairie group, the collective span of bloom 
 includes the entire growing season. 
    As already pointed out, the date of first bloom 
  in early marsh species is strongly affected by the 
  presence or absence of springs. 
    Some marsh plants have highly specific animal 
  relationships. Robert A. McCabe (unpublished) has 
  found that the alder flycatcher in the University 
  Arboretum uses the dry stalks of Angelica [160] 
  as a territorial perch, and the shredded bark of the